Welcome to AppFail
You last visited: never

Welcome to AppFail

Posted on 2009-06-24

Infrastructure Post

Infrastructure is all the various bits that hold a website up, like the piers of a bridge, or the legs of a chair. For your average blog or forum the most common failure is overloading the database, which happens when unoptimized queries lock tables for too long, or when a single database just cannot handle the volume of queries that are coming in. This results in the dreaded 'Too many Database Connections' error that is all too common on wordpress blogs submitted to digg. This is not actually the fault of wordpress, but of the poor choice in hosting provider by the blogger. A small investment in infrastructure will pay off every time. Another common problem is simply overloading the http servers capacity for concurrent connections. A single server can only answer a limited number of requests at any one time, and exceeding this limit generally causes an HTTP 503 error, or a connection timeout. The problem stems from that fact that you cannot just keep throwing hardware at the problem, a bigger server is still going to hit its limit eventually, and adding a second server, while it sounds easy, is not. How do you synchronize content, comments, and sessions across multiple servers? Now, you need to consider more than one database server, but how do you integrate new records added to separate database servers? The questions don't have easy answers, and the number of gotchas is endless. This is why it pays to have someone who has experience building and managing that infrastructure for you.

One of my pet peeves is when a favourite web comic is down because the newest comic has been submitted to digg, and I end up hammering refresh at a 503 error, or loading the page without the CSS or the images (the entire purpose of my visit). I understand a web comic doesn't generate a lot of revenue to pay for expensive infrastructure, but surely a more modern version of apache, or better yet, a fast httpd like nginx could service that load without breaking the site, or they could use a cloud solution like Amazon's Cloud Front service.

By: Allan Jude

blog comments powered by Disqus

Cuiusvis hominis est errare; nullius nisi insipientis in errore perseverare - Any man can make a mistake; only a fool keeps making the same one.

Digg Proof Hosting
The key to surviving Digg and Slashdot is Infrastructure. You can't get it from a regular web host, it requires experience. The High Load Hosting Experts at ScaleEngine can make your site thrive, and avoid having your site featured on AppFail.

Cyber Security Alerts

Page Generated in 15ms